DIY Ethernet shield (W5100)

Sorry to awake a 5 year old thread, but I guess such it's inevitable with such a niche discussion :slight_smile:

Dzhus, I'm really interested in putting together my own board incorporating a W5100 (see here to read about it). Basically an Arduino Nano with wired, PoE ethernet on board. The board's form factor would of course be different to the Nano, as it would definitely need to be bigger, but the idea is to keep it small.

I too have looked at the Freetronics and W5100 shield schematics, but I have so many questions about your project. I'll ask them here, as I couldn't find your contact details on this forum and your avatar isn't clickable!

SPI enable. Your first version just tied high, with a jumper to gnd. Then your second version has a transistor fed by SCS. The Arduino shield uses a 74LVC14D inverter which is according to this page discontinued. What is the purpose of this inverter? With your original version, I'm guessing this means the micro couldn't share the same SPI bus as the W5100, but with your version with transistor, it can? The Freetronics version has a 2N7002, and the schematic says "SPI MISO tri-state by SS/nCS. SPI D10_SS/nCS low = SEN high = SPI enabled". Haven't managed to fully understand what the 2N7002 does here as well.

Your schematic. Sorry I'm not familiar with the software you use, but the little red bits at the end of lines, what do they signify? I can see that some lines / wires are missing between connectors and the chip that are clearly supposed to be connected. I guess the labels can be used to match them? (The labels aren't exactly the same, but it's probably obvious)

1.8VD and 1.8VA. These rails marked on your schematic are connected along with a ferrite bead to W5100 pin V18. I'm guessing this is for supply conditioning, but what exactly does this do? What's the purpose of these rails? I can't see them referred to or linked to anywhere else in the schematic.

On the first version, can you explain the jumpers on SEN pin and /INT?

In your original post, you talk about "lower crystal" and "merging grounds". Can you explain further? What do you mean, lower frequency crystal? Did this cause problems for you, if so what problems? What lower crystal would you use instead? Can you explain the different grounds, e.g. I think I understand there were different ground planes for analog and digital, but I don't understand why. I presume there's only one supply? I'd love to hear a basic explanation about analog and digital ground planes, and different supplies, as I don't really understand the idea of an analog supply or a digital supply... I only understand the idea of supplying a single voltage to a chip.

Thanks in advance!